Gen3DSR: Generalizable 3D Scene Reconstruction via Divide and Conquer from a Single View

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ardelean, Andreea, Özer, Mert, Egger, Bernhard
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917972244692992
author Ardelean, Andreea
Özer, Mert
Egger, Bernhard
author_facet Ardelean, Andreea
Özer, Mert
Egger, Bernhard
contents Single-view 3D reconstruction is currently approached from two dominant perspectives: reconstruction of scenes with limited diversity using 3D data supervision or reconstruction of diverse singular objects using large image priors. However, real-world scenarios are far more complex and exceed the capabilities of these methods. We therefore propose a hybrid method following a divide-and-conquer strategy. We first process the scene holistically, extracting depth and semantic information, and then leverage an object-level method for the detailed reconstruction of individual components. By splitting the problem into simpler tasks, our system is able to generalize to various types of scenes without retraining or fine-tuning. We purposely design our pipeline to be highly modular with independent, self-contained modules, to avoid the need for end-to-end training of the whole system. This enables the pipeline to naturally improve as future methods can replace the individual modules. We demonstrate the reconstruction performance of our approach on both synthetic and real-world scenes, comparing favorable against prior works. Project page: https://andreeadogaru.github.io/Gen3DSR
format Preprint
id arxiv_https___arxiv_org_abs_2404_03421
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Gen3DSR: Generalizable 3D Scene Reconstruction via Divide and Conquer from a Single View
Ardelean, Andreea
Özer, Mert
Egger, Bernhard
Computer Vision and Pattern Recognition
Single-view 3D reconstruction is currently approached from two dominant perspectives: reconstruction of scenes with limited diversity using 3D data supervision or reconstruction of diverse singular objects using large image priors. However, real-world scenarios are far more complex and exceed the capabilities of these methods. We therefore propose a hybrid method following a divide-and-conquer strategy. We first process the scene holistically, extracting depth and semantic information, and then leverage an object-level method for the detailed reconstruction of individual components. By splitting the problem into simpler tasks, our system is able to generalize to various types of scenes without retraining or fine-tuning. We purposely design our pipeline to be highly modular with independent, self-contained modules, to avoid the need for end-to-end training of the whole system. This enables the pipeline to naturally improve as future methods can replace the individual modules. We demonstrate the reconstruction performance of our approach on both synthetic and real-world scenes, comparing favorable against prior works. Project page: https://andreeadogaru.github.io/Gen3DSR
title Gen3DSR: Generalizable 3D Scene Reconstruction via Divide and Conquer from a Single View
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2404.03421